3 What has to be managed in a project

MAIN AREAS OF ACTIVITY

Having spent considerable time in discussing what a project is, and the way in whichits management differs from that of a steady-state enterprise, it is now time to getdown to the subject proper. Furthermore, it is possible that some readers will havequickly skipped over the first two chapters in order to get to the meat of the subject.The purpose of this chapter is to set out the main areas of activity which have to bemanaged in a project. This is done with particular reference to construction, butthroughout the chapter separate comments are made to show how what is being saidwould apply to other situations. There are several distinct areas which have to bemanaged, and these are listed below. The following list is similar to that quoted byother authors (Turner, 1993) but has been extended and given different emphasis.

• Scope What is the project? How is it defined?• Procurement What contractual and organisational route will be used?• Planning and progress Analysing the project, setting a plan of action, and then

controlling progress.• Time What is the timetable, and how will it be enforced?• Cost What is the budget, and how will it be controlled?• Quality Determination of standards and their observance.• People Individual project managers and project teams.• Risk What are the risks, who carries them, and how can they be avoided or

minimised?• Project success/failure• Facilities The use of capital assets (e.g. buildings) and their maintenance.

It was stated above that these areas can be identified as being distinct, in that it isimportant that attention is paid to these primary aspects. This does not mean thatthey are independent of each other; they clearly interact in a number of ways. Forexample, it is well known that time and cost are each influenced by the other; if it isrequired to accelerate some work one of the simplest (but often not cheapest) meansof doing this is simply to work overtime. On the other hand if something arises tocause a delay in a project there may often be costs involved because workers are leftidle until the delay is overcome. This interaction of time and cost is an important

30

Chapter 3 WHAT HAS TO BE MANAGED?

aspect of project management and it is fully discussed in Chapter 8, along with the

link of quality with time and cost. Let us first look at each of the above topics in more

detail.

The scope of a projectIt may at first seem obvious that this can be simply stated as being 'to build a school','to launch a new product', 'to run an international conference' etc. This is, however,more like a concise mission statement since it does very little to define what theproject really includes. In the case of a building project it would be necessary todetermine the following and assign responsibilities.

(a) Who is the client? Is this the real client or simply an agent acting for theultimate owner to whom the building will be transferred on completion? It hasbeen common practice for all buildings to be used by government departmentsto be commissioned through an agency. Is the named client to be the ultimateuser? Many buildings are constructed for development companies who thenlease them to tenants on a long-term basis.

(b) Is it clear where funding for the project originates? Even where an agent ordeveloper is identified as the original client they may be obtaining finance froma bank or investment company who will be interested in monitoring wheretheir money is being spent.

(c) Is the project part of a larger strategy or programme of projects? If so how dothey tie up if at all?

(d) Is the client an 'intelligent' client? That is, are they in the habit of sponsoringprojects regularly, or is this an isolated event?

(e) Are the client and the contractor bound together by some form of partneringagreement?

Answers to these questions will determine how the scope is to be defined and how itrelates to other work.

Most building projects will be one-off and purpose-made, and will therefore needto be fully briefed and designed. It is necessary to determine whether these are to beincluded in the scope of the project under consideration, or whether they form part ofa separate project. One recent practice in use in some types of work has been toseparate the briefing and design from the actual construction; this has perhaps beendone to avoid some of the legal and liability arguments by separating responsibilityfor the two aspects and making it clear that design is the responsibility of thearchitect and engineer, while construction is the responsibility of the contractor. Thisis understandable but it has been found that it does not solve the problems of projectdelays, because it leaves grey areas which are not properly covered by the twoparties, leaving the client to cover them. However, this does lead to an interestingdiscussion of responsibility in project management which is left to Chapter 11.

The definition of the scope of the project is clearly very important. In the case of abuilding it will have to spell out carefully the physical site boundaries, possibleconnections with other buildings, the inclusion of all the appropriate services, the

31

CONSTRUCTION PROJECT MANAGEMENT

necessary financial and legal requirements and so on. Further coverage of these isgiven in Chapter 4, here we are simply concerned with the type of activity involvedand making the important point that the scope definition is an essential pre-requisiteof all the other stages of project implementation. The scope definition is largely aclear statement of what has to be done.

It is possible that the scope of the project will include not only the design andconstruction of the building, but also its furnishing and even its initial operation. Inthe case of a factory building it is now common practice for the project to include thepurchase and installation of all the production equipment. This may be extended toinclude the initial purchase of raw materials, the recruitment and training of staff andlabour, and even the setting up of transport and distribution systems for the product,together with all the computer hardware and software to manage the production unit.While this may not be a very common practice in the industrialised world it is oftenrequired for the establishment of a production plant in a developing country. Even inthe case of an 'intelligent7 client in a highly industrialised environment there will be aneed for the project management of a new production facility; it may not be part ofthe main contract and not within the scope of the building project, but all the stepslisted above will have to be managed, possibly by the client organisation itself in theform of a sub-project within a project.

The development of a new productThe preceding paragraph prompts the concept of a project which does not involvebuilding as such, but does include many of the features which make projectmanagement appropriate. Consider the case of an established manufacturingcompany which has decided to replace one of its existing products by a new one.Let us assume here that no building work is involved, and that all the actual work isto be carried out by the company itself. First the scope of what has to be included inthe project must be defined, possibly by asking if the following are within the projector not.

• Market research related to the new product.• Design of the new product.• Production methods to be used, i.e. the engineering of the product. (Does this

imply decisions on whether the new design will be constrained by existingproduction plant, and to what extent re-equipping will take place?)

• Sources of materials and components.• Checking of new and impending legislation relative to the product.• New standards within the industry.• Checking patents and licences.• Packaging and supporting operating instructions.• Distribution methods.• Changes in working practices within the plant.• Operative training, and perhaps recruitment or redundancy.• After-sales servicing.• Product launch and promotion.

32

Chapter 3 WHAT HAS TO BE MANAGED?

Some or all of these activities will be included and will have to be managed. They areall time-consuming and have costs, and there are many interactions between them.They form together a one-off exercise, namely a project, which lends itself to projectmanagement. This type of project will be more fully examined later in the book in thediscussion of a case study concerned with the development of a new product,together with the design and construction of a completely new plant to manufactureit, for Photo Products Ltd.

Procurement means, and organisational environmentWhen a new capital facility is to be commissioned and established, one of the firstthings that has to be done is to decide on the means that will be used to achieve it. Thisstage, which may be thought of as being pre-project, does actually form a veryimportant part of the project and has to be properly managed. There is an essentialstage to be undertaken in relation to business decisions about the size, location andgeneral purpose of the facility, backed by project appraisals both in terms of technicaland economic feasibility. These interesting and necessary studies form part of the detailof Chapter 4 on scope, but for the purposes of the present discussion let us assume thatwe are concerned with a project where all such reviews have been completed and thescope of the project has been defined. It is likely that the means of procurement will beproject-specific or at least industry-specific, and therefore it is appropriate to take asimple look at a number of cases. Given that we are concentrating on construction inthis book, let us examine a building project first of all. The most common traditionalpractice was for the client company to appoint a consultant architect to help with thebriefing of the project and then carry out preliminary designs for the purpose ofobtaining rough cost estimates, and in order to let the client see what form the buildingwould take. After a number of iterations of the design and costing, a decision toproceed would be taken and more detailed work would be put in hand. (This process isby no means as simple as is implied by this statement, and it will be further discussed inChapter 4.) Having got to the stage of an agreed outline design for the project, anumber of decisions have to be taken about the contractual method to be used incarrying out the detailed design and building construction.

In most buildings it will be necessary to employ specialist designers for thestructural engineering and services, e.g. heating and ventilating, and it has to bedecided whether these should be engaged by the client directly or by the architect.(The detail of this is included later in Chapter 11 dealing with forms of contract.)Mention of contract does however bring up one of the main decisions to be taken atthe start of a project, namely the contract form to be used. At this stage of the book itis only necessary to point out that the contract to erect the building can take any oneof a number of forms. Broadly contracts can be either lump sum where the contractorquotes a single fixed price for the work, or re-measured in which case the contractor'stotal contract figure will be based on the actual work completed, charged for at pre-agreed rates for each item of work. Within these two general forms there are manyother factors to be considered, e.g. will the contract be based on fixed rates or willsome measure of inflation be allowed for? Other forms of contract may include the

33

CONSTRUCTION PROJECT MANAGEMENT

detailed design stage, i.e. 'design and build', or 'turnkey'. Within the constructionindustry it is likely that civil engineering projects will be handled by engineers ratherthan architects, and there are correspondingly special contract forms for such work,e.g. the New Engineering Contract. The whole subject area of forms of contract is acomplex one in construction, and has become a special interest of the quantitysurveying profession, although this is a specialist group which does not exist in everycountry of the world. In the UK the quantity surveyors have been among the firstprofessions to take up project management, and given the complexity of manycontract forms this is understandable. It is of interest to note that some of the morerecently developed procedures have been prepared by project managers who arethemselves engineers (e.g. the New Engineering Contract and the British PropertyFederation System).

Some construction clients may themselves be big corporations who includeamong their staff building professionals, and as a result they do not need to engagespecialist consultants. There may still have to be decisions taken about the form ofcontract, and whether the work will be undertaken by the corporation itself with itsown regular or hired-in labour. Whatever method of procurement is chosen for aproject, it is easy to see that the early stage of a project where decisions on thesematters are being made calls for careful and effective management. If mistakes aremade in setting up contracts, these can be very expensive to overcome, whether interms of cost, time, or quality.

Installation of a management information system

This case of a non-construction project illustrates the same need for the means ofprocurement to be managed. A medium-sized manufacturing company with a longtradition and somewhat out-dated methods realises that it must establish a goodmanagement information system if it is to operate effectively and efficiently in future.Perhaps the first step in the project would be to engage consultants to study thebusiness and analyse its needs. This would correspond to the briefing stage of abuilding project. It is likely that this would produce a report proposing a detailedstudy of the company's procedures and its information needs, followed by a systemsdesign to provide for the means of meeting those needs. The systems design shouldbe in terms of the functions to be performed, and the outputs required, rather than aspecific software and hardware package. It is even possible (but unlikely) that asolution without computers could be proposed, and perhaps this is one of the mainproject decisions which has to be made. Another major decision is whether anexisting 'off-the-shelf computer package of software and hardware should be boughtand customised to the company's needs, or whether client-specific software should bewritten; it is unlikely that special hardware would be designed, except for specialattachments to production machinery. This decision can be a crucial one, partly interms of cost but more importantly in terms of delivery date and reliability. Manycomputer projects have run into real difficulty as a result of this decision, either bytrying in vain to modify an existing program which was not suitable, or converselyby grossly underestimating the difficulty of the task of writing client-specificsoftware, and the time and resources needed. Some of the most notorious failures in

34

Chapter 3 WHAT HAS TO BE MANAGED?

information systems have revolved around this problem, again showing the need tomanage the process properly, for example by pursuing more than one route for atleast part of the way if the consequence of failure is dramatic.

Other types of project will have their own procurement decisions to be made. Forexample setting up a new production facility for a new product will have two inter-related concurrent projects, namely the new product itself and the establishment ofnew plant for its manufacture. The product design may well be undertaken in-house,but may also involve some input from a design company. Who is going to undertakethe market survey, and how will this be fed back into the design of the product? Withregard to the establishment of the production line, will this be done by modifying anexisting line which currently produces an obsolescent product, and if so will there bea gap in production between the old and the new? Is this likely to be a problem, orcan it be overcome by stock-building? Who will make decisions about the selectionand purchase of new machinery? Is it feasible to modify the existing production line,and who will make this decision? Does the whole process of up-dating the productand plant require specialist advice, or will the whole project be handed over toanother company? There are clearly many questions to be asked about theorganisation of the changeover project, and these should be resolved at the outset.There are many projects of this type which have run badly, due to the lack of properdefinition of the organisational structure; who is producing what, who is makingdecisions, and who is carrying responsibility. Again there is a need for the propermanagement of the means of achieving the project on time, within budget, and up tothe specified performance. In some ways it is easier to measure the performance ofthis type of project, since this may be expressed in terms of a required output ofproduct measured in units per week. In the case of a building project some of theperformance criteria are long-term, e.g. withstanding gale-force winds which may notbe experienced for many years, or the intervals between exterior redecoration. Similardiscussions will take place on many different types of project, e.g. the development ofa specialised hospital department (Manton et al, 1991).

Planning and progressIt was said in Chapter 2 that the management of any production process could berepresented by the plan-measure-control cycle, and this applies equally well toproject management. In the case of a project however, its unique nature makesplanning both more difficult and more necessary. In the case of steady-stateproduction the production plan may consist of little more than what was done lastweek. Even where circumstances change, there is usually some good recentexperience upon which a plan can be based. In the case of a one-off project there islittle or no direct prior experience upon which the planner can work. At the sametime the fact that most projects are made up of a large number of interdependentitems of work makes it all the more important that reliable plans are made; it is notvery efficient just to let the project run ahead unplanned, in the way that a productionline may sometimes be allowed to continue, on the assumption that 'if it was OK lastweek then it is OK for this week'. There is the added complication that where the

35

CONSTRUCTION PROJECT MANAGEMENT

nature of the work on a project changes from week to week then different resourceswill be required. In fact one of the main differences between production and projectwork is that in the former the resources may be more or less static, whereas in thelatter the resources needed change very frequently. In order to cope with thisvariation in demands for resources it is essential that the work is carefully planned. Itcannot be too strongly emphasised that careful and detailed planning is one of theessential and most important aspects of successful project management. Therealisation of this and the absence of appropriate techniques in the past led to thedevelopment of network planning methods, which form the main content of Chapter5. Unfortunately there was a feeling in the early days of project management thatnetworks would themselves solve all the problems of controlling projects, but thishas proved to be grossly over-optimistic. It is now realised that network methodsprovide very useful tools, but as in all other activities they must be properly used andbacked with a range of other methods. Before studying these methods in detail it isuseful to see how the progress on a project can be measured.

Models in the measurement of progress

Measurement of the output from a steady-state production process is usually fairlyeasy by counting or measuring the output in terms of units per week. There is adifficulty in the case of project work because the actual nature of the work will oftenchange on a day-to-day basis, and furthermore the output does not lend itself tobeing counted in terms of uniform units of production. Over a period of years anumber of ways of measuring project progress have been tried, and most of themhave been found to have short-comings. It is useful to describe some of these in orderto understand why networks have provided a good basis both for planning and formeasuring progress. One way of doing this is to consider a simple example which isnot specific to any one industry or type of project.

Here we consider only the implementation stage of a project, assume that all thepreliminary decisions and design have been completed, and an order has been placedfor the work to be done. The value of the project is £ l million and the plannedproject duration is ten months. There is a strong temptation to measure progress infinancial terms, partly because one of the main parameters of the project is its value,and partly because most organisations will have most of their records in terms ofmoney, e.g. wage records, purchases of materials, hire charges for equipment, andreceipts from clients. This may have led to the adoption of the very simplistic yetwidely-used basis of measurement on a project which assumed that the total projectcost could be divided by the duration of the project to give a notional value of workper month which must be completed. In the example quoted above this would meanthat the total cost of £ l million would be spread over the ten-month period to give anotional value of £100 000 per month for the work to be done. This is illustrated inFig. 3 which shows the cumulative cost of the work over the period of the project.The planned cost is shown as a straight line rising uniformly over the ten-monthperiod. The actual cost of the work achieved can then be plotted on the same basisand a comparison made between work planned and work completed, as measured bythe costs incurred to date.

36

10

g| 6

Chapter 3 WHAT HAS TO BE MANAGED?

Cumulative value

v90% cost

Assumed uniformspend rate

Cumulativecost

80%'time

9 10

Time: months

Fig. 3. S-curve showing cumulative cost and value

There are two problems here. It is often the actual costs of work done rather thana true measure of what has been achieved that is recorded, simply because costinformation is readily available. The rate of activity on a project is seldom uniform.The use of cost data can be misleading, because they will include the cost of anyabortive work and the correction of faults etc.; what should really be measured is thereal value of acceptable work that has been completed (in the case of a buildingcontract this would be based on an interim valuation of work).

A build-up of the rate of working presents the second difficulty, and requires moredetailed explanation. On any project it is likely that initially there will be relativelyfew people working on it, but that the team will build up to a maximum number andthen decline towards the end of the work. (For a computer installation this wouldmean a few systems analysts at the start, building up to hardware specialists,programmers, applications writers, and installation teams, and than tailing off at theend to the small commissioning team. In the case of a building the initial small teamwould be the excavator operators, then concrete gangs, bricklayers, carpenters,electricians, plumbers, with probably only the painters and cleaners left at the end.) Itis not only the direct labour costs which follow this pattern, since it is likely that thequantities of materials used and equipment hired will roughly follow the number ofpeople working on the project, indicated by the lines on Fig. 4. This well-understoodphenomenon is represented by what is known as an 'S-curve' when plotted as acumulative figure, as shown on Fig. 3 above and further discussed in Chapter 6. It iswhen we compare plan with performance that the importance of the two lines onFig. 3 can be seen.

37

CONSTRUCTION PROJECT MANAGEMENT

Number of workers

• — Costs per day

Time

Fig. 4. Spend rate and labour strength during project

Cumulative-cost curves are applied in the simple example discussed above, andFig. 3 shows both the uniform cumulative curve (straight line) and the non-uniformline (S-curve). Consider initially that we have used the straight line as the plannedrate of work, but that the actual rate of work has followed the S-curve, and assumethat the project has been completed on time. There will be three points at which thetwo lines coincide, at the start and finish of the project and somewhere around itsmid-point. At each of these points a check on the diagram will indicate that theproject is 'on programme', but let us consider more closely the eight-month stage ofthe project (i.e. 80% of the duration) where the straight-line target will indicate aproject value at that time of £800 000 (i.e. 80% of the project total value).Examination of the S-curve will however show that the project value at that timeought to be around £900 000 if the work is to be completed on schedule. If in realitythe project value at the eight-month stage is only £800 000 then it can be seen thatthe project is behind schedule and is following a cumulative value curve which willoverrun by one or two months. This of course is what happens in practice; manyprojects seem to be going well for much of their duration, and then for someunknown reason near the end are found to be behind programme. There are manyreasons for this occurrence among them being the inherent optimism of traditionalmanagers (see Chapter 10), but also the fact that any activities which are not urgenttend to get left 'until there is more time', but then bunch up at the end and run intoresource problems (see Chapter 9). The conclusion from this simple fictitious case sofar is that the measurement of cost or value against a straight-line assumption is not asound basis for the measurement of progress on a project.

The concept of an S-curve is useful in terms of project financial control but it doeshave limitations as a means of measuring actual progress. This can be illustrated byreference to another simple case, that of a new product which will be packaged in atotally new way, and distributed through a new dealer network. The project hasmany strands to it, and it may be that the new packaging and distribution systemsmake very good progress and create real value for the project. They will, however,have done nothing to advance the design and manufacture of the product itself whichmay well be falling behind programme. In a similar situation a building project may

38

Chapter 3 WHAT HAS TO BE MANAGED?

appear to be going well because its first few months' value is up to the S-curve target;closer inspection may reveal that much of the value created has been due to theconstruction of a high perimeter wall around the whole site, valuable in itself butdoing nothing to advance the actual construction of the building. The finalconclusion of this discussion of monitoring progress is that money does not providea suitable unit of measurement, except in the sense of controlling finance. Otherindustries do not use money as their main measure of output, the oil industryproduces barrels of oil, the motor industry counts numbers of vehicles, coal-minesproduce tonnes of coal, and service industries such as transport record passenger-miles. One of the challenges to project management has been to devise a means ofmeasuring progress when there is no consistent unit of output such as tonnes orpassenger-miles, and which does not use money as its base. The method adopted hasbeen to create a model of what has to be done and then compare actual performancewith the model. The model usually takes the form of a network as this can representvery well the interactions between all the activities which make up a project. Themodels exist either as drawings or as a series of linked statements in a computer.They serve a number of purposes, starting with providing a means of thinking aboutthe project, then acting as the basis of a plan to check time, cost and feasibility, andfinally acting as the basis for comparing actual progress with the plan. Models inmany situations are also used for experimentation where actual testing is either toodifficult, expensive or dangerous (e.g. the ultimate load testing of a bridge). Networksas models permit the testing of activity sequences when clearly actual tests would bevirtually impossible in one-off project work. The foregoing discussion has led to thearguments leading to the development of networks in project planning and control.The detail of how this is done is the subject of Chapter 5.

Time managementMuch of the previous section covering the management of planning and progress isconcerned with the management of time, and indeed the time element is apredominant consideration in networks. At the start of this section it may seemsomewhat simplistic to ask what we mean by time, but it is useful because there are atleast two aspects which must be considered. There are effectively two types of time incommon use in projects, and there is confusion between them. First there is time in thesense of resource use, e.g. the worker-hour. There is a long tradition in manyindustries, especially heavy mechanical engineering, to measure work in terms of man-hours. This is because it permits the planning of the use of resources which may befixed in total, and also is closely related to costs, especially labour costs which form afairly high proportion of total costs in the labour-intensive project industries. Theman-hour is a useful means of stating the work content of an activity, often a truerwork measure than the cost in money terms since it is not subject to inflation and doesnot carry overhead or profit elements; it does nonetheless suffer from some variabilitydue to changes in working conditions and working methods. It is sufficiently useful asa base measure for many operations in construction and heavy engineering to justifythe publication of 'standard' man-hours for a wide range of tasks.

39

CONSTRUCTION PROJECT MANAGEMENT

The other use of time is in the sense of elapsed time or clock time, and this has tobe distinguished from work content. The difference can perhaps be best illustrated byan example. A particular operation in the execution of a systems analysis of acomputer application is estimated to require 30 man-days of work by a systemsengineer; this means that it could be completed by 1 person in 30 working days or in15 working days by 2 people. In the first case the elapsed time would be 30 days andin the second case only 15 days; the work content would be 30 man-days in eachcase. It may seem tedious to emphasise this point but the author has seen manyinstances of confusion between the two uses of time. One of the topics covered inChapter 8 concerns the ways in which we may seek to reduce the elapsed time on aproject, but for the present it is worth giving it some attention. The above simpleillustration indicated that it might be possible to reduce the elapsed time required foran activity by doubling the number of people working on it. Depending upon thenature of the work it may or may not be possible to make a further similar reductionby allocating another person to the task, making 3 in all, in an attempt to reduce theduration to 10 days. This cannot go on for ever and it would probably be quiteimpossible to seek to complete the work in 1 day by using 30 people. It may well bethe case that even with 3 people the work cannot be totally shared out and that itwould take say 12 days of elapsed time (rather than 10) to complete the activity. Thisof course means that the actual work expended on the task would have increasedfrom the original of 30 man-days to 36 man-days, with the consequential increase incost. An alternative to increasing the number of people working on a task from 2 to 3would have been to ask the 2 people to work overtime by increasing their workingweek from 40 hours to 60. This might just succeed in finishing the task within anelapsed time of 10 days and a work content of 30 (standard) man-days. A clearunderstanding of the two uses of time is essential for anyone taking up projectmanagement, especially if they come from a background of contract estimating inwhich all work has been evaluated in terms of man-hours. In most projects the clientwill primarily be interested in the duration or elapsed time of a project, or perhapseven more truthfully in its completion date. While the supplier or contractor is alsointerested in completion by the agreed date, there is no doubt that the man-hourcontent of achieving that date is often of equal importance.

The preparation of a working plan may be undertaken on any of the many waysof planning and measuring work on a project; these are primarily based on time, andseveral of them are looked at in detail in Chapter 5. One of the traditional ways hasbeen to represent separate tasks in the form of a bar chart, but as will be shown thisdoes have some shortcomings. The same also applies to a simple list of target dates.It was because of these that network methods were evolved, and their use is nowwidespread in project work. It must be pointed out at this stage however thatnetworks do not provide all the answers, and there are many instances wherebarcharts and/or lists of dates are the most appropriate. Time planning and controlare essential to good project management, and it is useful to compare the planning ofa project by analogy with its design. The technical design of a new structure willinvolve many sequential stages.

40

Chapter 3 WHAT HAS TO BE MANAGED?

(a) The initial concept.(b) The briefing of its form and specified performance.(c) The loading and other requirements.(d) Stress analysis, other calculations and sizing of members.

(e) Consideration of alternative designs and making choices.

All of these will be completed before the final stage:

(/) Preparation of the working drawings.The planning of the erection of the structure is in effect the design of the erectionmethod, and will go through a corresponding series of numbered steps, as follows.

(a) Analysis of what has to be done, in terms of activities.(b) The logic sequence of the work.(c) The elapsed time and resource needs of each activity.(d) The calculations of project duration and its adjustment.(e) Consideration of alternative methods and making choices.

All of these will be completed before the final stage:

(/) The preparation of the plan of action for completing the work.

In much the same way that it is not normal practice to go directly to the detailedworking drawings at the outset of the design stage, it is not good practice to gostraight to the barchart as the plan of action for the construction. Some plannersattempt this and usually find it unsuccessful. The underlying theme in both designand planning must be.

(a) Define what is to be achieved.(b) Analysis and understanding.(c) Data input.(d) Detailed calculations for achieving the target.

(e) Comparing alternatives and choosing the most appropriate.

All of these must be completed before the final stage:

(/) The stated plan of what is to be done and how.The final plan of action for the implementation of a project may exist in the mediumof drawings, written statements or computer outputs. It may take the form ofnetworks, barcharts, date lists or even written words. Barcharts have the positiveadvantage of giving a pictorial representation of progress to a time scale and arereadily understood. They are less useful at the earlier analytical stage, for which someform of network is more appropriate, and is really essential. The essence of this couldperhaps be stated: networks for analysis, barcharts for presentation. This is not agolden rule, but is a guide to good practice.

Cost managementThe traditional way of controlling costs in industry is to set a budget forecast andthen compare actual performance with the budget. It is then usual to highlight any

41

CONSTRUCTION PROJECT MANAGEMENT

significant variances from budget, seek the cause and take remedial action whereappropriate. This approach is entirely consistent with the plan-measure—control cyclepreviously described in Chapter 2 (Fig. 2.). There are often two broad approaches tocost control; one is more global, i.e. to plan and then measure the actual inputs to theprocess being considered, for example the consumption of raw materials, utilisationof labour, use of equipment, energy consumption etc. This approach will cover totaluse of an input item rather than cost per unit of output, but it does give good overallcontrol and can quickly alert management to major problem areas such as power orother energy leaks, wastage or theft of materials or components, various types offraud and so on. This type of control is unlikely to pick up smaller but still importantdiscrepancies in terms of efficient use of materials, operator time loss because of lackof parts etc. One of the ways to track this type of problem is to use the secondapproach to cost control, a system of 'unit costs' by which all the cost inputs to aprocess are recorded on a week-by-week basis and then any trends can be identifiedand acted upon. Such unit costs would be separated into a number of categoriesincluding main material inputs, labour, equipment use, energy and so on. Anextension of this procedure is not simply to compare one week's cost figures withthose of the preceding weeks, but to establish a 'standard cost' which is what the unitcosts should be if everything goes well. The object of this is to avoid figures whichare consistently poor and therefore do not show up in comparisons. Regular checkingagainst a calculated standard will highlight such continuously poor performance. Oneof the problems with the use of the standard cost method is the difficulty inestablishing the initial figure for inputs, and this complex task is the responsibility ofwork study personnel. It can be fraught with difficulties especially where it is linkedto the setting up of targets for bonus schemes, and can itself be a costly exercise. It isthrough this approach however that detailed studies of the production methods inuse can make significant improvements by examining carefully all aspects of aprocess, e.g. the physical layout of the workplace, the storage of components, the useof working jigs in place of measurements, the automation of simple repetitiveprocedures, and many other similar items.

The management of costs in projects is difficult. The above description of costmanagement in steady-state production is given not as a means of managing costs inprojects, but to show that there is a contrast between the two types of industry in theway they carry out cost management; it is necessary that this difference isappreciated if good control of costs is to be exerted. There has been much time andeffort wasted on projects trying to undertake costing exercises on the principlesoutlined above which were designed for steady-state production. As an example theauthor was engaged in a study to record and compare the weekly costs of fixing andstriking shutters for a large concrete wall; over a period of several weeks apparentvariations in cost were found to be due to inconsistencies in recording and majorchanges in weather conditions. These disturbances took some weeks to resolve, bywhich time the wall was complete, and the only lesson to be learned was the futilityof the exercise on a one-off job. Subsequently work was done on other sites, andfrom this some general principles for the design and use of shutters were established,but it did not produce a good weekly cost control system. The unique nature of

42

Chapter 3 WHAT HAS TO BE MANAGED?

projects and the lack of continuity of any individual item of work does mean that the'conventional' approach to cost control is not really applicable. As a consequence arange of other techniques have been evolved to manage costs, and these aredescribed in Chapter 6. Significant work has been done in this area by J.R. Turner, asdetailed in The Handbook of Project-Based Management.

Quality managementThis simple two-word phrase has recently become one of the new management areasof great interest and the subject of many books, articles, seminars and courses. It hasbeen brought about by competition among manufacturers who now seek to expandtheir markets by improving quality rather than simply competing on price. Theelectronics and motor vehicle industries have made great advances in terms ofquality, and there are now major efforts being made to bring the benefits of goodreliability and quality into the project-based industries. Quality is the subject ofChapter 7 where again the special case of one-off project work is described. The mostimportant point to make here is that the slogan often used in quality management'get it right first time' is nowhere more true than in projects, simply because there isseldom a second time. In manufacturing it may be desirable to get things right firsttime, but it is usually possible to throw away the first unit of production even thoughit may cost money to do so. This is a luxury which is not possible in project work.

A fundamental point in the consideration of quality is the definition of what wemean by the word, and it is important to make this clear. There are at least twoaspects which we must think of separately. In common usage we think of high orlower quality goods when what we really mean is their standard. For example, wemay think of a Rolls Royce car as being of high quality compared with the lowerquality of a mass-produced car such as a Ford or VW, when what we really mean isthat the standard of specification is higher. The meaning of quality in productionterms is the degree to which every unit of production meets the standards whichhave been set. In relation to motor vehicles, customers have every right to expectthat the Ford or VW vehicle will meet the stated standards and will be reliable, and infact these and other motor manufacturers have been able to achieve very goodreliability rates. It could well be argued that in mass-produced vehicle production it iseasier to maintain high standards because so many of the processes are automatedand therefore not subject to the risk of human error that exists with the hand-builtcar. This does provide a link with project quality, since the production of individuallybuilt vehicles is very close to being project work where a rather more meticulousscrutiny of every operation becomes necessary. The distinction between the twoaspects of quality that we have identified is that the first can be thought of as 'qualitydecision', and the second as 'quality attainment'. In this text the phrase 'qualityassurance' has been avoided since it has a very specific meaning in some contexts.

Another important area concerning quality in projects is the interaction with timeand cost and this is pursued in Chapter 8. There is often a trade-off between thespecified standard for a product and its cost, either in relation to the materials used,the allowable tolerances, or other attributes. There is sometimes also a trade-off with

43

CONSTRUCTION PROJECT MANAGEMENT

the time required, especially in the case of projects. This is particularly true whensome slightly defective work is detected and a decision has to be taken to correct it atthe expense of a delay in the project. This is often a 'catch-22' situation to whichthere is no easy answer, but one which does need to be addressed, as is done inChapter 8.

The management of peopleIt is in this area that project management probably has most in common with themanagement of other types of activity. Clearly any undertaking that employs asignificant number of people will have to use a range of skills concerning themanagement of people, including communication, leadership, delegation, recruitment,selection and training; there are many good texts on these and it is not relevant todiscuss them at length here. There are aspects however where project managementdoes emphasise particular aspects of people management, and these are worthclarifying. Chapter 10 is entitled 'People in project management', but is really mostlyconcerned with the characteristics and skills required to be a project manager, and theways in which teamwork plays a major part in project management. Our immediateconcern here is the management of other people in the project, and this includes notonly the operatives executing the work but also the client's staff, professionalconsultants, government officials, bank representatives and so on. In Chapter 2, thediscussion of jobbing work pointed out that much of the influence over what is donein a project is literally in the hands of the operative or individual worker. Furthermorethe work that has to be done will change from day to day, and it cannot be assumedthat just because today's work is going well that tomorrow's will also be carried outproperly. A simple and perhaps trivial example of this is the case where a brickworkgang on Monday starts the building of a house wall; by Tuesday it is clear that theyare careful workers with all courses laid truly to line and level, proper bondingobserved, all joints clean, very little waste, and with good progress being made. Theforeman is confident that they are reliable and is too busy to look at their work againuntil Thursday, by which time the wall has been well built, but with windowopenings in the wrong place! How often are errors like this found on a buildingproject and what does that tell us about the need for close control in projects, withdetailed supervision of work? More will be said about this in Chapter 7, but it isworth stating here that the consequences of such errors in projects can be very greatas the cost of correcting or re-working can exceed the initial cost of the work in bothmoney and time. There is a great need for clear instructions in delegation, goodsupervision and appropriate training. This contrasts with steady-state productionwhere there is still a need for supervision but the same instructions will apply forseveral days at a time, and the only real management task is to ensure that thingskeep running smoothly. The management of cost, time, and people is furtherdiscussed by Wearne (1989).

It is not only in connection with operatives that close supervision in projects isneeded. For example it is common in many project-based industries to employspecialist consultants or sub-contractors to carry out part of the work. This again

44

Chapter 3 WHAT HAS TO BE MANAGED?

requires a special approach. For example, a manufacturing company may sub-contractsome of the components of its product and can call upon the supplier to submitsamples for checking; it then has the ability to reject these samples or any subsequentitems which do not come up to the required standard. In the case of project work thismay not be so easy, as the sub-contract work is being done on the project site, and inreal project time. For example, a specialist machine-tool company may be installing anew production line and chooses to sub-contract some minor building works toaccommodate the new equipment. The fact that the building works are small incontent and value does not reduce their importance in getting the project complete; asmall roof leak which does not become apparent for a few weeks, by which timedelicate plant has been installed, can prove to be very serious. Most project managerswill be able to relate similar horror stories.

Project managers will also have to learn how to deal with designers and otherprofessionals who frequently regard themselves as the most important people in aproject and therefore have to be handled very carefully. Even more difficult is theclient's representative who may be extremely demanding and will often point outwho is paying for the project and therefore who has the right to call the tune. Itdoes not always go down well with clients to point out the terms of appointmentof the project manager and/or the terms of the contract, even though these maygive the project manager complete authority over the matter in question. This typeof people management requires great diplomacy as well as good communicationskill.

Risk managementThis is another of the phrases which has come into the vocabulary of managementliterature in recent years, and like so many of these concepts means different things todifferent people. For example, the insurance industry sees itself as the home of riskmanagement, but while insurance makes a major contribution to the subject there aremany other aspects. Essentially risks arise because there is an element of uncertaintyin everything we do, and also a variability in the outcome of every process. What weare seeking to do in risk management is to manage variation and uncertainty. Theactual concept of risk is a measure of the consequences of the occurrence of theunexpected, rather than the chance or probability that it will happen. There are manyrisks associated with project work, only some of which can be covered by insurance.On a building site it may be possible to insure against flood or severe wind damage,but it is unlikely that compensation could be obtained for the slowing down of workdue to high winds or prolonged rain; these are not disasters but merelyinconveniences. There are commercial risks to be considered in relation to the pricesof components and raw materials, uncertainties about labour rates that will prevail,the possibility of new legislation which will impact on the project, whether the clientwill want modifications but is not prepared to accept a delay, and many others whichare examined in Chapter 13, and are discussed in an introduction to the subject of riskmanagement (Thompson and Perry, 1992).

CONSTRUCTION PROJECT MANAGEMENT

Project success/failureAt the end of a project it should be possible to make an assessment of the extent towhich it has been a success or failure. In order to do this it is helpful if criteria areestablished at the planning stage which can be used as the basis of the assessment.Some of these criteria are easily identified.

• Was the project completed by the target date?• Was the project completed within the budget?• Has the completed project met all the performance standards set?

Answers to these questions will to some extent be dependent upon the point of view,whether it is that of the client, designer, contractor/supplier, or the actual user. Mostparties to the project will be interested in completion on time, but the attitude to costwill vary. The client will be keen to keep the total cost to a minimum, but thecontractor/supplier will wish to maximise the profit margin between actual cost ofwork and the price paid by the client. The designer will wish to see a good job welldone, and both the client and the final user/occupier will be anxious to see thatquality and performance are as specified.

Some projects will fail to meet their expectations in some respects, and some maybe total failures, even to the point of being abandoned before completion. Generallyall parties will celebrate success, and are not keen to publicise failure, but it is useful toundertake an analysis of project performance, so that lessons may be learned for thefuture. Major projects and their shortcomings are likely to be discussed publicly.

Facilities managementThis subject is not so much an essential part of project management but rather anadd-on optional extra, and indeed many project managers will never have anycontact with it. The subject has been studied for some time, and previously wasreferred to as asset management, but this caused confusion with the financialoperators who simply saw assets as locations to invest money with a view toobtaining reward, either in terms of dividends or capital appreciation. Assetmanagement was really about making the best use of buildings, plant and equipment,i.e. assets which were in the possession of the organisation for the purposes of theirproduction. This could mean the best utilisation of a fleet of aircraft, the optimalproduction schedule for a factory, or the most effective use of building space in ahospital or university. Asset management has now come to be known as facilitiesmanagement and has become part of project management partly because projectmanagers are the people who install the new facility or modify the old one. It iscommon practice for large organisations to have a plant strategy or estate strategy tocreate, maintain and operate their stock of plant and buildings; the preparation of thisstrategy and its implementation and updating is often made the responsibility ofproject managers within the organisation.

46

Chapter 3 WHAT HAS TO BE MANAGED?

DISCUSSION OF PROJECT MANAGEMENT

This brings us to the end of the general discussion of the subject of projectmanagement and the ways in which it compares with and differs from themanagement of manufacturing. This has been given at some length in order to give agood understanding of the fact that we are dealing with a different subject area andhave at times to think differently. The following chapters deal with the Tiow-to' ofproject management, namely the techniques and methods which have beendeveloped to help practise the discipline. Further reading on the general applicationsof project management can be found in various books (Cleland and King, 1988; Reiss,1992; Stall worthy and Kharbanda, 1984). A number of major projects which give amost interesting context to the subject are also described (Morris and Hough, 1987;Leech and Turner, 1990; Smith (ed.), 1995). There is a thorough and perceptivehistory of project management, illustrated by reference to many well-known projectsby Morris (1994).

Terminology of project management

Throughout this book and most of the relevant literature there are many specificwords and phrases which are used to denote precise concepts. Where it is importantto be accurate these terms have been defined, but in addition a language of projectmanagement has evolved and is explained (Cleland and Kerzner, 1985), and is also setout in a number of British Standards (BS 4335, BS 6046 and BS 6079).

BRITISH STANDARD BS 6079: 1996, Guide to Project Management

BS 6079 was published at a time when this book was in the early stages of printing. Itis therefore only possible to give some brief discussion of the document itself, ratherthan comment upon the use to which it has been put in industry. First impressionshowever would indicate that if its general messages are taken to heart by clients andcontractors/suppliers then it will indeed improve the performance of projects. This isits stated objective, since it starts from the position that 'the overall track record ofBritish organisations in managing projects, including takeovers, leaves much to bedesired'. This is a view which can be disputed however, since within recent years theperformance on many projects has been very good, especially where projectmanagement has been properly applied. There are still several much-publicisedfailures, many of these due to the fact that thorough project management has notbeen put in place, especially by clients. BS 6079 addresses this point and its statedobjective is 'to help managements to deal with these situations more efficiently andeffectively'.

The new British Standard was prepared by a committee, many of whose membersare related to building and civil engineering or other heavy construction, but alsoincluding aviation, electronics, telecommunications, defence, quality and others.Participating organisations are listed alphabetically, with the APM first; reading thedocument also indicates that the influence of the APM has been a very important and

47

CONSTRUCTION PROJECT MANAGEMENT

leading one. The introduction to BS 6079 gives a good one-page statement of thedevelopment of project management and its place in modern industry. It is also clearthat it provides a set of guidelines for newcomers to the management of projects andan aide memoire for those with some experience. It is not a prescriptive standardwhich lays down mandatory requirements, which was a fear that had been expressedby a few people before it was published.

There is a useful list of definitions of the terms used in project management. Allthe terms used in this book and most of those used in other texts are consistent withBS 6079 which also includes some additional definitions which are of note. One ofthese is 'task' which is defined as 'the smallest indivisible part of an activity when it isbroken down to a level best understood and performed by a specific person',although this does seem to be too restrictive; in many projects there may be taskswhich have to be performed by a team. Another definition is 'statement of work'(SOW), which is 'a document stating the requirements for a given project task'. Giventhat the task is stated to be the smallest indivisible part of an activity there will be avery large number of tasks and consequently a large number of SOWs. This couldlead to a great deal of paperwork which might become a distraction from the real jobof getting on with the project activities.

The content of BS 6079 is a very good general statement of project management,and it provides a list of project phases very similar to the one given in Fig. 5 (page52) in the next chapter of this book, including the frequently ignored final phase of'disposing of residual assets (liabilities)'. The standard also deals briefly with theimportant area of project organisation. Considerable attention is given to the projectmanagement processes of planning, measurement and control; this includes workbreakdown structure, task dependency and criticality, costs, earned value, cash flow,deliverables, milestones, resources, quality, configuration management, risks,procurement, subcontracting, project personnel and their required skills, project lifecycle and several other related topics.

Overall BS 6079 is a good concise statement of what has to be done, rather thanhow to do it. In under 50 pages it gives a broad introduction to how the managementof projects should be approached, and as such it will provide useful guidance to thesenior management of both clients and contractors/suppliers who have notpreviously made proper use of project management; it is not a superficial document.At the operating level these guidelines remain relevant but will need to besupplemented by much more detail.

48